README.md

---
license: cc-by-nc-4.0
library_name: femr
tags:
- healthcare
- femr
- medical
extra_gated_prompt: "You agree to all terms outlined in 'The EHRSHOT Credentialed Health Data License' (see https://shahlab.stanford.edu/ehrshot_license). Access requires a verified CITI training certificate using the same process outlined by PhysioNet (see https://physionet.org/about/citi-course/) Please provide proof via the verification URL, which takes the form https://www.citiprogram.org/verify/?XXXXXX. You agree to not use the model to conduct experiments that cause harm to human subjects." 
extra_gated_fields:
 Full Name: text
 Email: text
 Affiliation: text
 CITI Certification Verification URL: text
 I agree to all terms outlined in 'The EHRSHOT Credentialed Health Data License': checkbox
 I agree to use this model for non-commercial use ONLY: checkbox
---

# MOTOR-T-Base

This is a 143 million parameter autoregressive foundation model pretrained on 2.57 million deidentified EHRs from Stanford Medicine. 

This is the model from [(Steinberg et al. 2023)](https://arxiv.org/abs/2301.03150).

As input, this model expects a sequence of coded medical events that have been mapped to Standard Concepts within the [OMOP-CDM vocabulary](https://ohdsi.github.io/CommonDataModel/index.html). The model generates representations of patients which can then be used for downstream prediction tasks.

## Model Details

### Model Description

- **Developed by:** Shah lab @ Stanford University
- **Funded by:** Stanford Healthcare
- **Shared by:** Shah lab @ Stanford University
- **Model type:** MOTOR [(Steinberg et al. 2023)](https://arxiv.org/abs/2301.03150)
- **Language(s) (NLP):** Electronic health record codes
- **License:** CC-BY NC 4.0
- **Finetuned from model:** N/A -- trained from scratch

### Model Sources

- **Paper:** [MOTOR: A Time-To-Event Foundation Model For Structured Medical Records](https://arxiv.org/abs/2301.03150)

## Uses

This model is intended to generate representations for patients based on the structured data within their electronic health record. 
These representations are ideally used for time-to-even-modeling, but can also be used for other downstream tasks such as predicting diagnoses, detecting anomalies, or doing propensity score matching for causal inference.

### Direct Use

You will likely want to tune the model for your downstream use case. 

### Out-of-Scope Use

This model is for research purposes only. It is not for use in any real-world decision making that impacts patients, providers, or hospital operations.

## Bias, Risks, and Limitations

This model was trained on a corpus of 2.57 million patients from Stanford Medicine. 
The model will thus reflect the patterns of how care is delivered at Stanford Medicine, in addition to the racial and socioeconomic makeup of Stanford Medicine's patient base. 
This model may not generalize well to other hospitals and demographic mixes. 

## How to Get Started with the Model

We recommend getting started by looking at our tutorial repository: https://github.com/som-shahlab/motor_tutorial

## Training Details

Full training details are provided in our accompanying paper, [MOTOR: A Time-To-Event Foundation Model For Structured Medical Records](https://arxiv.org/abs/2301.03150).

### Training Data

The model is trained on 2.57 million patients from the [Stanford Medicine Research Data Repository (STARR)](https://academic.oup.com/jamiaopen/article/6/3/ooad054/7236015), which contains EHR data from both Stanford Health Care (primarily adult care)
and Lucile Packard Children’s Hospital (primarily pediatric care). 
The dataset contains only structured data (i.e. no clinical text or images) and covers demographics (e.g. age, sex, race), diagnoses, procedures, laboratory results, medication prescriptions, and other coded clinical observations. 
The data is formatted according to the [Observational Medical Outcomes Partnership Common Data Model (OMOP-CDM)](https://ohdsi.github.io/CommonDataModel/cdm53.html). 
All data that we work with is deidentified.

### Training Procedure

We train our model using an time-to-event pretraining objective, i.e. predict the time until a particular code appears in a patient's timeline.

#### Preprocessing

We use the [FEMR](https://github.com/som-shahlab/femr/tree/main) Python library for data preprocessing.

#### Training Hyperparameters

* Learning rate: 1e-5
* Context window size: 496
* Internal dropout: 0
* Layers: 12
* Hidden dimension: 768

## Evaluation

We evaluate this model on Stanford data, see [MOTOR: A Time-To-Event Foundation Model For Structured Medical Records](https://arxiv.org/abs/2301.03150). 

## Technical Specifications

This model uses the MOTOR architecture from  [MOTOR: A Time-To-Event Foundation Model For Structured Medical Records](https://arxiv.org/abs/2301.03150).

## Citation

**BibTeX:**
```
@misc{steinberg2023motor,
      title={MOTOR: A Time-To-Event Foundation Model For Structured Medical Records}, 
      author={Ethan Steinberg and Jason Fries and Yizhe Xu and Nigam Shah},
      year={2023},
      eprint={2301.03150},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}
```
## Model Card Authors

Ethan Steinberg, Michael Wornow

## Model Card Contact

Ethan Steinberg (ethan.steinberg@gmail.com)